*4.1. Histopathology*

Histopathology is one of the strongest predictors of sperm retrieval as it provides a direct snapshot of the testicular architecture [23]. However, performance of a testicular biopsy solely for diagnostic purposes is not routinely recommended because of its invasiveness. In addition, a diagnostic biopsy samples only a small section of the testicular tissue, so its predictive value is limited. Diagnostic biopsy is suggested at the time of retrieval attempt to document the condition treated and rule out pathologic processes such as intratubular germ cell neoplasia. Additionally, testis biopsy may also be performed to differentiate maturation arrest from normal production in men with normal volume azoospermia, normal serum follicle stimulating hormone (FSH) concentrations, palpable vas deferens, and normal testicular volume [24].

Histopathologic subtype has been correlated to sperm retrieval rate, and those with Sertoli cell only syndrome histopathology have lower sperm retrieval rates compared to those with maturation arrest or hypospermatogenesis patterns [25]. As expected, the presence of mature spermatozoa is a strong predictor of sperm retrieval [26]. Unfortunately, despite the utility of testicular histopathology, it often is not available prior to sperm retrieval procedures, and is of limited value when no spermatozoa are seen [27]. For example, men with a diagnostic biopsy showing Sertoli cell only syndrome are expected to have sperm production elsewhere in the testis in at least 37% of cases. At our institution it is standard that pathology reports include all histologic patterns present in a testis biopsy, as even small foci of spermatogenesis are correlated with successful sperm retrieval. Pathology reports that state only the predominant or most severe histopathology are highly unlikely to reflect the likelihood of sperm retrieval in men with NOA.

#### *4.2. Testis Size*

In a meta-analysis by Corona et al., men with testis size >12 cc had higher rates of sperm retrieval, however, sperm retrieval was still possible in small volume testes (<8 cc) [28]. Our observations are the opposite; that men with larger testes are more likely to have obstructive azoospermia, as suggested by Schoor et al., and that men with NOA have similar sperm retrieval chances, regardless of testis volume [27]. With an effective mTESE search for rare foci of sperm, the testis volume or FSH level (which reflects overall testicular function) cannot predict the region of best function/sperm production inside the testis. Other meta-analyses have demonstrated limited predictive value of testicular volume even when testis biopsy histopathologic patterns were also used in the analysis [29]. Overall, these data sugges<sup>t</sup> that testis size should not be considered a factor to exclude a patient from an attempt at sperm retrieval. In fact, in our experience, sperm can be routinely retrieved even in testes less than 2 cc in volume [30].

#### *4.3. Serum Follicle Stimulating Hormone Levels*

Serum FSH concentrations have been suggested, in some isolated reports, to predict sperm retrieval in conventional TESE [31]. Other studies have reported FSH levels to inversely correlate with the number of germ cells present and stages of spermatogenesis [32]. However, while high serum FSH levels may provide a more global representation of the level of spermatogenic dysfunction within a testis, there still may be small foci of spermatogenesis that can be identified and retrieved during mTESE [33]. Therefore, we do not recommend using baseline serum FSH concentrations as a preoperative predictor of sperm retrieval in NOA men.
